Otomatik depolama ve boşaltma sisteminin modellenmesi ve optimizasyonu
Modeling and optimization of automated storage and retrieval system
- Tez No: 735129
- Danışmanlar: PROF. DR. CEVAT ERDEM İMRAK
- Tez Türü: Yüksek Lisans
- Konular: Makine Mühendisliği, Mechanical Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2022
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Lisansüstü Eğitim Enstitüsü
- Ana Bilim Dalı: Makine Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Konstrüksiyon Bilim Dalı
- Sayfa Sayısı: 167
Özet
İnternetin ve internet alışverişinin tüm dünyada yaygınlaşmasıyla beraber insanlar ihtiyaç duydukları ürünlere daha hızlı ulaşabilir hale gelmiştir. Teknolojinin gelişimiyle artan bu ihtiyaç karşısında şirketlerin ürünleri zamanında teslim edebilmeleri gerekmektedir. Bu durum özellikle e-sipariş için geçerlidir. Zira internet tüketicileri genellikle siparişler için bir günlük döngü süresine ulaşan küçük miktarlarda bir ila iki mal sipariş ederler. Pazarın bu ihtiyacını karşılayabilmek için firmaların bir mal siparişinde yer alan tüm aşamaları kısaltması, artan rekabet ortamında hayatta kalabilmesi için gereklidir. Bu tez kapsamında bir mal siparişindeki önemli bir yeri tutan depolama sistemlerinden Otomatik Depolama ve Boşaltma Sistemleri (ODBS)'nin performansının arttırılması için çalışma yapılmıştır. Literatür araştırması yapılarak ODBS'lerin performanslarının arttırılmasına yönelik çalışmalar incelenmiştir. İncelemeler sonucunda üç ana yöntem gözlemlenmiştir. Bu yöntemler sırasıyla mal alma/bırakma politikalarının iyileştirilmesi, kolon tasarımının iyileştirilmesi ve ODBS pozisyon kontrolünün iyileştirilmesi şeklinde sıralanabilir. ODBS performansının arttırılması için yapılan önceki çalışmalar genellikle alma/bırakma politikasının optimizasyonu üzerine yoğunlaşmıştır. Bu nedenle bu tez kapsamında daha hızlı bir ODBS tasarımı için kolon tasarımı alternatifleri sunulmuş ve alternatifler arasından seçilen kolon tasarımının pozisyon kontrolü yapılarak ODBS performansının artışı simüle edilmiştir. Tez kapsamında öncelikle yüksek hız ve ivme değerleri belirlenmiş ve bu değerlere uygun bir kolon tasarımı belirlemek için çalışma yapılmıştır. Yapılan çalışma sonucunda 6 adet kolon tasarımı üzerinde durulmuş ve bu kolon tasarımlarının birbirleriyle ağırlık ve mukavemet göz önüne alınarak karşılaştırılması yapılarak optimum kolon tasarımı belirlenmiştir. Seçilen kolon tasarımının pozisyon kontrolü kutup yerleşimi yöntemi kullanılarak yapılmıştır. Pozisyon kontrolü sonucunda elde edilen grafiklerde seçilen kolon tasarımının üzerinde ataletten kaynaklı olarak oluşan titreşimlerin çok düşük olduğu ve tasarımın seçilen hız ve ivmelerde hareket etmeye uygun olduğu değerlendirilmiş ve ODBS'nin performansını arttırdığı simulasyon sonuçları ile doğrulanmıştır.
Özet (Çeviri)
With the spread of the internet and internet shopping all over the world, people have become able to access the products they need faster. In the face of this need, which increases with the development of technology, companies need to be able to deliver their products on time. This is especially true for e-ordering. Because internet consumers often order one to two goods in small quantities that reach a one-day cycle time for orders. In order to meet this need of the market, it is necessary for companies to shorten all the stages in a goods order, in order to survive in an increasingly competitive environment. Within the scope of this thesis, a study has been carried out to increase the performance of Automatic Storage and Retrieval Systems (ASRS), which is one of the storage systems that has an important place in a goods order. Studies to increase the performance of ASRS's were examined by making a literature review. As a result of the investigations, three main methods were observed. These methods can be listed as improving the goods pick-up/drop-off policies, improving the column design and improving the ASRS position control. In the studies carried out to improve the first method, the policies of receiving / leaving goods, the frequency of ordering the products is determined first and a model is created to determine the shelf placement of the products. When the layout is made according to the model, the most frequently ordered products are located on the nearest shelves. Finally, with this model, improvements in delivery times are shown. Only one article was found in the literature on improving the column design, which is the second method. A composite column design is proposed in the found article. The reason for this recommendation is that the vibration damping rate of the selected composite material is higher than the damping rate of metals such as steel or aluminum. Thus, the time required for vibration damping is reduced. Only one article was found within the scope of improving the position control, which is the third method. In this article, first of all, the dynamic model of ASRS is extracted. The resulting model was controlled using the gain scheduled control method. As a result of this study, it has been shown that vibration can be prevented and the movement time can be shortened by position control. Previous work to improve ASRS performance has generally focused on optimizing the pick/drop policy. For this reason, within the scope of this thesis, column design alternatives are presented for a faster ASRS design and the increase in ASRS performance is simulated by checking the position of the column design selected among the alternatives. In this thesis, the basic ASRS design is emphasized. The column structure of the basic ASRS designs is divided into two main groups as single column and double column according to the weight of the load carried. While double column ASRS is used for loads heavier than 500 kg, single column ASRS design is preferred for lighter loads. Basic ASRS's operate on a single corridor and can carry one or two loads simultaneously. Since it was decided that the ASRS to be designed within the scope of the thesis should carry load weighing 50kg, it was decided to design a single column, unit load-bearing foundation ASRS. Before starting the decided ASRS design, the features of the ASRS systems that are widely used in the market were examined. As a result of this examination, the speed and acceleration values of the system of the major company are listed and higher values than these values are selected within the scope of the thesis. Then, a basic ASRS model was drawn and the mass values of the carrier cars on the model were determined. The loads acting on the column were calculated using the determined parameters and a study was carried out to determine a column design suitable for these loads. As a result of the study, 6 column designs were emphasized. In the design of the first column, a rectangular profile with a length of 300 x 200 x 12000 mm and a thickness of 6 mm was used. The material of the profile used was chosen as St37 (S235). According to this selected material type, the total weight of the column design was calculated as approximately 550 kg. In the design of the second column, 2 rectangular profiles with a length of 200 x 100 x 12000 mm and a thickness of 3 mm and 42 square profiles with a length of 50 x 50 x 330 mm and a thickness of 3 mm were used. The material of all profiles used in the column design was chosen as St37 (S235) and the total weight of the column design was calculated as approximately 384 kg. In the design of the third column, 4 square profiles measuring 50 x 50 x 12000 mm and 2 mm thick and 168 square profiles measuring 50 x 50 x 330 mm and 2 mm thick were used. The material of all profiles used in the column design was chosen as St37 (S235) and the column weight was calculated as 287 kg. In the fourth column design, a rectangular profile measuring 202.3 x 101.6 x 12000 mm and a thickness of 6.35 mm was used. The material of the profile used in the column design was chosen as Al 7005 and the column weight was calculated as 120.6kg. In the design of the fifth column, 4 square profiles measuring 110 x 110 x 12000 mm and 2 mm thick and 120 square profiles measuring 110 x 110 x 380 mm and 2 mm thick were used. The material of the profiles used in the column design was chosen as Al 7005 and the column weight was calculated as 195.6kg. In the design of the sixth column, 4 square profiles measuring 90 x 90 x 12000 mm and 2 mm thick and 120 square profiles measuring 90 x 90 x 380 mm and 2 mm thick were used. The material of the profiles used in the column design was chosen as Al 7005 and the column weight was calculated as 166.81kg. All column designs were analyzed with the Finite Element Method (FEM) according to the load values determined before the design phase, and Von-Mises stress values and deflection values were found. By comparing these column designs with each other considering weight and strength, the fifth design was chosen as the optimum column design. Since the design phase has been completed, the position control phase has been started. While performing the position control, a dynamic model was obtained by using the article found in the literature research. Spring and damping coefficients in the dynamic model were obtained by using FEM. The state-space equation of the system in continuous time was obtained by establishing the dynamic model equations and system parameters in the MATLAB R2017b student version. By using the continuous time state-space model, the discrete time state-space model is also derived. Many controller designs such as PI (Proportional Integral), PID (Proportional Integral Derivative), LQR (Linear Quadratic Regulator) and pole placement have been tried for position control of the system model. Since the best performance is obtained from the pole placement controller among the controller designs, other controller designs are eliminated. The prerequisite for using the pole placement controller design is that the system is both observable and controllable. Therefore, it was ensured that these features were provided before the use of the pole placement method. Since it has been observed that the pole placement method can be used, it has been passed to the stage of moving the roots of the system to the desired locations. At this stage, firstly, the roots of the discrete-time state-space system are found. After a few trials, the roots of the found system roots were determined. In order to determine the gain value to be used in the pole placement method, the Ackermann formula was used and the gain value was determined. The values found up to this stage were transferred to Simulink (R2017b). The controller design established in Simulink processes signals at two different times. The input signal is taken in discrete time, multiplied by the gain value, then converted to continuous time and fed into the system matrix. The position control is done by converting the values at the output of the system matrix to discrete time and giving feedback to the input part. From the position control results, it has been observed from the graphics that the system goes to the desired position stably in a very short time. As a result of the thesis study, it was seen that the geometric design of the column and the material selection had a direct effect on the weight of the column. Parameters of the load carried by ASRS, such as weight, velocity and acceleration values of ASRS are directly effective in column design and position control. It has been observed that giving the input signal in discrete time while performing position control and determining the optimum time step while creating discrete time equations increase the controller performance. As a continuation of this study, column model and controller design can be suggested by following the same steps for ASRS designs with higher weights. In addition, the ASRS examined in this study is driven only by the engine on the lower car. In the follow-up studies, the effects of adding an engine on the upper car on the column design, ASRS performance and controller design can be examined.
Benzer Tezler
- Experimental investigation of the different shared storage assignment policies in the end-of-aisle asır systems
Koridor sonu otomatik depolama boşaltma sistemlerinde değişik paylaşımcı depolama politikalarının deneysel incelenmesi
ZAFER BOZKURT
Yüksek Lisans
İngilizce
1995
Endüstri ve Endüstri MühendisliğiOrta Doğu Teknik ÜniversitesiDOÇ.DR. CANAN SEPİL
- Yüksek verimli bir otomatik depolama ve boşaltma sisteminin tasarımı ve gerçekleştirilmesi
Design and realization of a high efficient automatic storage and retrieval system
BARIŞ KABLAN
Yüksek Lisans
Türkçe
2024
Mekatronik MühendisliğiKocaeli ÜniversitesiMekatronik Mühendisliği Ana Bilim Dalı
PROF. DR. HÜSEYİN METİN ERTUNÇ
- Depolama işlemleri için labview tabanlı bir robotik sistemin geliştirilmesi
Labview for storage operations developing a robotic system based
BÜŞRA ALTUN KELEŞ
Yüksek Lisans
Türkçe
2019
Mekatronik MühendisliğiSakarya Uygulamalı Bilimler ÜniversitesiMekatronik Mühendisliği Ana Bilim Dalı
PROF. DR. SİNAN SERDAR ÖZKAN
- Otomatik depolama ve boşaltma sisteminde optimum çekme politikasının belirlenmesi
Defining optimal retrieval policy in an automated storage and retrieval system
AYLA ÖZTÜRK
Yüksek Lisans
Türkçe
2011
Endüstri ve Endüstri MühendisliğiGazi ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. EDİZ ATMACA